Research On The Performance Of AI Listed Companies Based On Traditional DEA And Three-stage DEA Models

Artificial Intelligence(AI)is the core driving force of the Fourth Industrial Revolution and a strategic technology that leads the future.The main driving factor for the development of AI is the AI enterprise,but the current comparative research on the performance of AI enterprises in high-tech enterprises and the impact of environmental factors on the performance of AI enterprises are relatively lacking.Therefore,this paper will build an AI innovation performance and comprehensive performance evaluation index system,and use all listed high-tech enterprises from2016 to 2019 as samples to study the performance of AI enterprises.The traditional DEA model is used to compare the corporate performance level,and the three-stage DEA model is used to evaluate the corporate performance level after removing the non-management influence factors.According to the performance comparison between high-tech listed enterprises and AI listed enterprises based on the traditional EDA model,the corporate innovation performance efficiency and comprehensive performance efficiency have always been at the average level among high-tech listed enterprises in the past four years,and the number of best-performing AI listed enterprises has been 0.Judging from the performance evaluation results of AI listed enterprises in the three-stage model,before and after the adjustment,the performance of corporate innovation performance is "big at two ends and small at the middle",and there is a large gap in performance level.A long period of enterprise establishment is not conducive to improving the utilization efficiency of enterprise total assets,and regions with high economic development level and high education level are not conducive to the utilization efficiency of enterprise innovation input cost.The overall performance of the company is higher and the gap between the levels is relatively small.The overall performance of the adjusted AI listed enterprises shows a significant increase in the innovation performance of the enterprise,and the gap in the overall performance level is further narrowed.The longer the enterprise is established and the higher the level of regional opening to the outside world,the higher the efficiency of the input variables;the higher the level of regional education,the higher the efficiency of the use of total assets and R&D personnel,and the lower the efficiency of the use of R&D inputs;government subsidies aren't conductive to the total assets and R&D personnel usage efficiency.Finally,according to the research conclusions,the following suggestions are put forward.In order to improve the performance of AI enterprises,the resource allocation mode of high-tech enterprises on the frontier of optimal efficiency should be used for reference to improve the efficiency of resource allocation.We will establish an industry-university-research collaborative innovation system,build a collaborative innovation system for universities,research institutions,and enterprises,so as to improve their capability for independent innovation.The enterprise will be set up in areas with high level of external development and abundant senior talents to improve the utilization efficiency of external resources.Give full play to the guiding role of the government,promote new incentive policies and guiding policies,and optimize the environment for enterprise development.